Lines Matching refs:methods

91 set iterative methods and preconditions for solving the linear systems,
93 information on the solver methods available in TAO.
172 documentation for individual methods for more details.
409 TAO fully supports matrix-free methods. The matrices specified in the
416 (`PCSHELL`). In other words, matrix-free methods cannot be used if a
417 direct solver is to be employed. Details about using matrix-free methods
855 methods available in TAO for solving these problems can be classified
862 3. Function, gradient, and Hessian evaluations – Newton Krylov methods:
869 trust-region methods will likely perform best. When a Hessian evaluation
880 TAO features three Newton-Krylov algorithms, separated by their globalization methods
1112 the linear system of equations. For the conjugate gradient methods and
1448 the Nash or Steihaug-Toint conjugate gradient methods or the generalized
1458 gradient methods and generalized Lanczos method, this preconditioner
1598 Dai-Yuan method. These conjugate gradient methods can be specified by
1680 Before introducing specific methods, we will first define two projection
1709 separated by their globalization methods: projected line search (BNLS),
1715 The BNK class of methods use an active-set approach to solve the
1750 method. Trust-region conjugate gradient methods (`KSPNASH`,
1865 methods. These methods can be specified by using the command line
1869 preconditioning for these methods, and it is controlled by the flag
1874 Some individual methods also contain their own parameters. The
1875 Hager-Zhang and Dou-Kai methods have a parameter that determines the
1896 All methods can be scaled using the parameter `-tao_bncg_alpha`, which
1901 diagonal scaling as an option for its CG methods, turned on with the
1903 (regular) and inverse Broyden methods are developed, controlled by the
2064 trust-region method (TAOBQNKTR). Other first-order methods such as
2134 interior-point methods such as PDIPM, the Hessian matrix tends to be
2477 and GPCG methods may not solve the subproblem. Therefore, a bounded
2682 theoretical properties of such methods are detailed in the
2775 TAO also contained two active-set semismooth methods for solving
2776 complementarity problems. These methods solve a reduced system
2940 Convergence of a solver can be defined in many ways. The methods TAO
2942 These methods include absolute and relative convergence tolerances as
3011 For the quasi-Newton family of methods (`TAOBQNLS`, `TAOBQNKLS`,
3036    methods. The solvers can obtain this function and derivative
3207 objects. These methods include adding vectors together and computing an
3208 inner product. A full list of these methods can be found in the manual